- Email: [email protected]
Dietary patterns and colorectal cancer
Clinical Nutrition xxx (2016) 1e5
Contents lists available at ScienceDirect
Clinical Nutrition journal homepage: http://www.elsevier.com/locate/clnu
Original article
Dietary patterns and colorectal cancer Reema F. Tayyem a, *, Hiba A. Bawadi b, Ihab Shehadah c, Lana M. Agraib d, Suhad S. AbuMweis d, Tareq Al-Jaberi e, Majed Al-Nusairr f, Kamal E. Bani-Hani g, Dennis D. Heath h a
Department of Nutrition and Food Technology, Faculty of Agriculture, The University of Jordan, Amman 11942 Jordan Human Nutrition Department, College of Health Sciences, Qatar University, P.O. Box: 2713, Doha, Qatar c Chief Gastroenterology Division, King Hussein Cancer Center, Jordan d Department of Clinical Nutrition & Dietetic, The Hashemite University, P.O. Box 150459, Zarqa 13115, Jordan e Jordan University of Science and Technology, Jordan f Chief Gastroenterology Division, Prince Hamza Hospital, Jordan g Faculty of Medicine, Hashemite University, Jordan h Cancer Prevention and Control Program, Moores Cancer Center, University of California, San Diego, La Jolla, CA 92093, USA b
a r t i c l e i n f o
s u m m a r y
Article history: Received 8 February 2016 Accepted 28 April 2016
Background & aims: Dietary pattern and lifestyle have been reported to be important risk factors in the development of colorectal cancer (CRC). However, the mechanism of action of dietary factors in CRC disease is unclear. The aim of this study is the examination of several dietary choices and their potential association with the risk of developing CRC. Methods: Dietary data was collected from 220 subjects who were previously diagnosed with CRC, and 281 control subjects (matched by age, gender, occupation and marital status). The data was collected between January 2010 and December 2012, using interview-based questionnaires. Multivariate logistic regression was used to estimate the relationship between dietary choices and risk of developing colorectal cancer. Results: Factor analysis revealed three major dietary patterns. The first pattern we identified as the “Healthy Pattern”, the second was identified as “High Sugar/High Tea Pattern” and the third as “Western Pattern”. In the Healthy Pattern group we found a 10.54% variation in food intake, while the intake variation was 11.64% in the Western Pattern. After adjusting for confounding factors, the Western Pattern food choice was found to be significantly associated with an increased risk of developing CRC (OR ¼ 1.88; 95% CI ¼ 1.12e3.16). The results for the Healthy and High-Sugar/High Tea Patterns showed a decrease, but the statistic was not significant for the risk of CRC development. Conclusion: The Western Pattern of dietary choice was directly associated with CRC. The association between the dietary food choice in the Healthy and High-Sugar/High Tea Patterns and colorectal cancer needs further study in our Jordanian population. © 2016 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.
Keywords: Colorectal cancer Dietary pattern Healthy dietary pattern Western dietary pattern
1. Introduction In colorectal cancer (CRC) disease, several well-known dietary and non-dietary risk factors have been implicated. Some of those factors are high consumption of red meat and processed meat; low
* Corresponding author. E-mail addresses: [email protected] (R.F. Tayyem), [email protected] (H.A. Bawadi), [email protected] (I. Shehadah), [email protected] (L.M. Agraib), [email protected] (S.S. AbuMweis), tmrjaberi@hotmail. com (T. Al-Jaberi), [email protected] (K.E. Bani-Hani), [email protected] (D.D. Heath).
fiber intake; alcohol drinking; obesity; and a sedentary life style [1]. Additionally, genetic susceptibility [2], tobacco smoking [3], and exposure to environmental carcinogens, were found to promote proliferation and malignant transformation of CRC cells [4]. Previous studies have focused on the effects of a single food item or a nutrient on lowering risk of CRC incidence [5]. However, the association of a single food item or food group with the risk of developing CRC may not be valid because of the presumption that each single food item or nutrient has an isolated effect [6]. A dietary pattern in food choice is defined as a combination of the dietary components (food items, food groups, nutrients, or both) used to summarize elements of the total diet or the major features of the
http://dx.doi.org/10.1016/j.clnu.2016.04.029 0261-5614/© 2016 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.
Please cite this article in press as: Tayyem RF, et al., Dietary patterns and colorectal cancer, Clinical Nutrition (2016), http://dx.doi.org/10.1016/ j.clnu.2016.04.029
2
R.F. Tayyem et al. / Clinical Nutrition xxx (2016) 1e5
food choices for the population under study [7]. The descriptive summary of the dietary pattern has been used in nutritional epidemiology to explain and assess the overall dietary experience, by suggesting that the synergistic effects of the variety of dietary and non-dietary factors can be used to explain the relationship between diet and health [8]. In general there are two dietary patterns; “Healthy” and “Western”. The healthy dietary pattern is largely characterized by a greater intake of fruits, vegetables, grains, and a lower intake of sweets, red meat, and processed meat, this dietary pattern is considered to be associated with lowering the risk of developing CRC. Alternatively, the Western dietary pattern, reported to contain more meat, highly processed food, potatoes, refined carbohydrates, and much lower in vegetables and dietary fiber, has been reported to increase the risk of developing CRC [7,9]. The numbers of studies that have examined the possible effect of dietary patterns and CRC disease in the developing countries are few. However, we previously reported on the association between macro- and micronutrients consumption and the risk for colorectal cancer among Jordanians [10]. In this study we aimed to examine the major dietary patterns and their possible association in the development of CRC in a Jordanian population, using factor analysis or Principal Component Analysis. 2. Materials and methods 2.1. Study population and methods A detailed description of the method has been reported elsewhere [11]. Briefly, a case-control study of 501 participants was conducted from January 2010 to December 2012; with 220 diagnosed with CRC disease, and 281 controls (a total of 248 males and 253 females). Patients diagnosed with CRC disease (cases) were recruited from five Jordanian hospitals specializing in oncology diagnosis and treatment. The Control group was recruited from outpatients, hospital personnel, and visitors. Controls were matched with cases for age, gender, occupation and marital status. The study protocol was approved by Institutional Review Board Committees from all hospital groups. Control subjects were excluded if any first- or second degree relatives were diagnosed with CRC. The following inclusion criteria for controls were used: Jordanian nationals aged 18 years or older, ability to communicate clearly and verbally, free of any type of diagnosed cancer, diabetes mellitus, liver disease and rheumatoid arthritis. For inclusion in the diagnosed CRC cancer group, subjects must have received their diagnosis less than 1 year prior to the time of the first interview. The exclusion criteria for this group included those who were considered “critically ill”, such as an in-patient at any facility and those who were unable to communicate verbally and clearly. Written informed consent was obtained from all subjects before their interview. 2.2. Data collection Socio-demographic information (age, marital status, household income, education, occupation and tobacco usage); medical history as well as dietary data were collected by trained research assistants using interview-based questionnaires. The comprehensive medical data included the participant's full medical history to confirm the status of CRC diagnosed subjects and disease-free subjects. A validated Arabic quantitative Food Frequency Questionnaire (FFQ), adapted from the Diet History Questionnaire (DHQ I) of the National Cancer Institute of the United States of America [12] was used for dietary assessment. The FFQ questions tracked the information on the dietary history of study participants prior to CRC
diagnosis, and to confirm the dietary habits of control participants. A period of one year prior to the diagnosis date was selected, to reflect seasonal variation in some food types. A constant dietary pattern for the period was noted in our participants, with most of the participants suggesting this pattern existed for at least five years. A qualified dietitian asked participants, during face-to-face interviews, how frequently, on average, during the past year they had consumed one standard serving of specific food items in nine categories (<1/month, 2e3/month, 1e2/week, 3e4/week, 5e6/ week, 1/day, 2e3/day, 4e5/day, or 6/day). An answer in the affirmative resulted in additional questions related to frequency and amount of food consumed. If the participants' dietary pattern did not include a food type, then related questions were skipped. Food lists in the modified FFQ questions were classified based on types of foods: 21 items of fruits and juices; 21 items of vegetables; eight items of cereals; nine items of milk and dairy products; four items of beans; 16 items of meat such as red meat (lamb and beef), chicken, fish, cold meat, and others; four items of soups and sauces; five items of drinks; nine items of snacks and sweets; and 14 items of herbs and spices [12]. For better portion size estimation food models and standard measuring tools were used. Dietary intakes were analyzed using dietary analysis software (ESHA Food Processor SQL version 10.1.1; ESHA, Salem, OR, USA) with additional data on foods consumed in Jordan. The 7-day Physical Activity Recall (PAR), developed by Sallis et al. (1985) was used to measure physical activity level. 7-Day PAR is a structured interview that depends on participant's recall of time spent engaging in physical activity over a seven day period [13]. Participants were asked to respond to a PAR question based on the way they used to behave prior to being diagnosed with CRC. The number of hours spent in different activity levels were obtained and converted into metabolic equivalents (METs). The total physical activity MET minutes per week was obtained by summing the METs and then performing categorical analysis (inactive, minimally active, or health enhancing physical activity active) [13]. Body weight (measured to the nearest 0.1 kg); height (measured to the nearest 0.5 cm); and body mass index (BMI) were taken and calculated as prescribed by Lee and Nieman [14]. 2.3. Statistical analyses Statistical analysis was performed with SPSS IBM-20 software. The significance level was set at p 0.05. For descriptive statistics, mean ± standard deviation (SD) and percentages were used. T-tests evaluated the differences between cases and controls in continuous variables, and Chi-square was used to detect differences among categorical variables. Dietary patterns were derived using Principal Component Analysis (PCA), form factor analysis. The food items in the FFQ were separated into 22 food groups, based on their similarity of nutrient content and culinary usage or their reported relationship with cancer [15] (Table 1). Kaiser-Meyer-Olkin (KMOtest) and Bartlett's test of sphericity were used to assess suitability for using factor analysis for this exercise. As suggested by Safari et al. (2013) [16] sampling adequacy and inter-correlation of factors were supported by KMO value > 0.632 and Bartlett's test of sphericity < 0.001, respectively. Communality index was assessed to indicate the variance in each food group being tested by the analysis [16]. Factors were retained based on an eigenvalue of >1 for screen plot. Then Varimax rotation was applied to review the correlations between variables and factors [16]. Food groups with absolute factor loadings >0.20 were considered to have contributed significantly to the pattern. Cases and controls received an individual factor score for each identified pattern [17]. Potential confounders (with age, sex, BMI, physical activity level (MET-min/ week), total energy intake, occupation, education level, smoking,
Please cite this article in press as: Tayyem RF, et al., Dietary patterns and colorectal cancer, Clinical Nutrition (2016), http://dx.doi.org/10.1016/ j.clnu.2016.04.029
R.F. Tayyem et al. / Clinical Nutrition xxx (2016) 1e5 Table 1 Food groups used in the dietary pattern analysis. Food group
3
Table 2 Selected characteristics of the participants' study. (Tayyem et al., 2016).
Food items
Processed meat Red Meat Fish Poultry Egg Dairies Tea Coffee Fruits
Sausages, hamburger, mortadella, turkey Beef meat, ground meat, lamp meat Tuna, any type of fish Chicken Fried eggs, boiled eggs Milk, yogurt, cheese, ice cream, labaneh. Black tea, green tea Coffee Watermelon, melon, apple, apricot, fig, nectarine, peach, pear, Citrus fruit, date, kiwi, grape, strawberry, banana, grape fruit, alovera, other fruits Tomato Tomato Carrot Carrot vegetables Spinach, lettuce, mixed vegetable, stew vegetables, eggplant, green squash, local vegetables, pepper, cucumber, garlic, kinds of cabbage, root vegetables, other vegetables Legumes Bean, chickpea, split pea, lentil, other cereals Fried potato Fried potato Boiled potato Boiled potato Total grains Bread, rice, macaroni Snacks Biscuits, puff, chips Nuts Peanut, almond, walnut, pistachio, hazelnut, roasted seeds Sweets and desert Cakes, cookies, chocolate, Arabic sweet, candy Sugar Sugar, sugar cube Mayonnaise Mayonnaise Soft drink Carbonated drinks
marital status and family history (beyond the second degree) for the CRC participants) were chosen based on reported risk factors for CRC [18,19], including the Cancer Prevention Study II [19]. 3. Results Cases and controls were matched for several parameters, including age, sex, occupation and marital status. Characteristics specific to our subjects were published previously [11] and shown in Table 2. The mean age for controls and cases was 51.3 ± 11.1 and 52.9 ± 11.7 respectively, while mean BMI was 29.1 ± 5.6 and 27.7 ± 6.2 for controls and cases, respectively. The Factor loading matrix analysis showed three major dietary patterns. These three distinct dietary patterns, labeled “Western Pattern”, “Healthy Pattern” and “High-Sugar/High Tea Pattern”, were extracted using the above-mentioned factor analysis procedure. The three dietary patterns explained 32.26% of the total variance in food intake. The first pattern was heavily loaded with processed and red meat, total grains, coffee, fried potato, nuts, poultry, sweets and desserts. This group of food was identified as the “Western” dietary pattern. The Western dietary pattern was responsible for 11.64% of the total variance. The second pattern, labeled “Healthy” dietary pattern, had a high amount of fruits, vegetables, dairies, tomato, carrot, and nuts. While, the third pattern identified as “High-Sugar/High Tea Pattern”, had food group that was significant due to the higher amounts of sugar and tea, and obtained results had a variance of 7.99%. The cumulative variance was about 30% (Table 3). After adjusting for confounding factors, an increased risk of developing CRC was positively associated with the Western dietary pattern (OR ¼ 1.88; 95% CI ¼ 1.12e3.16), while an insignificant decreased risk of developing CRC was observed in the Healthy Pattern food choice group (OR ¼ 0.93; 95% CI ¼ 0.56e1.53) as shown in Table 4. 4. Discussions This case-control study investigated several dietary patterns and their potential association with the risk of developing CRC. The
Age (mean ± SD) Sex n (%) Male Female BMI (mean ± SD) BMI Category Under weight (<18.5) Normal (18.5e24.9) Over weight (25e29.9) Obese 30 Marital status n (%) Married Single Divorced Widowed Occupation n (%) Yes No Smoking n (%) Smoker Non-smoker Family history of CRC n (%) Yes No Other health problem n (%) Yes No Education n (%) Illiterate primary and secondary Diploma and BSc MSc and PhD MET n (%) Inactivea Minimally Activeb HEPA activec
Control (n ¼ 281)
Case (n ¼ 220)
51.25 ± 11.12
52.89 ± 11.67
132 (47) 149 (53) 29.06 ± 5.6
116 (52.7) 104 (47.3) 27.71 ± 6.15
2 (0.7) 52 (18.8) 122 (44.2) 100 (36.2)
5 (2.5) 57 (27.9) 83 (40.7) 59 (28.9)
248 (88.3) 17 (6) 1 (0.4) 15 (5.3)
199 (90.5) 5 (2.3) 3 (1.4) 13 (5.9)
100 (35.6) 181 (64.4)
69 (31.4) 151 (68.6)
53 (18.9) 227 (81.1)
37 (16.8) 173 (78.6)
101 (36.5) 176 (63.5)
84 (38.5) 134 (61.5)
119 (42.7) 160 (57.3)
83 (37.7) 136 (61.8)
11 (3.9) 137 (49.1) 113 (40.5) 18 (6.5)
17 (7.7) 107 (56.4) 86 (39.1) 10 (4.5)
131 (52.2) 53 (21.1) 67 (26.7)
121 (56.3) 37 (17.2) 57 (26.5)
BMI: Body Mass Index. a Inactive: not fitting in “Minimally Active” or “HEPA active”. b Minimally Active: at least 600 MET per week. c Health Enhancing Physical Activity: HEPA active: more than 3000 MET per week.
Table 3 Factor loading matrix of food groups for Healthy and Western dietary patterns. Food group Egg Total grains Red meat Fish Coffee Soft drink Processed meat Sweets and desert Mayonnaise Nuts Snacks Chicken Legumes Fruits Carrots Vegetables Tomato Dairies Fried potatoes Boiled potatoes Tea Sugar Total variance
Western Pattern 0.512 0.491 0.490 0.451 0.449 0.431 0.428 0.400 0.388 0.388 0.369 0.350 0.292
0.212 0.335
11.64%
Healthy Pattern
High-sugar/ high tea pattern
0.274
.237 0.212 0.264
0.655 0.652 0.650 0.556 0.441 0.370 0.306
10.54%
0.240 0.231 0.865 0.836 7.99%
Absolute factor loading values <0.20 for the three patterns were excluded for simplicity. Bold values represent foods with highest contribution to each patterns.
Please cite this article in press as: Tayyem RF, et al., Dietary patterns and colorectal cancer, Clinical Nutrition (2016), http://dx.doi.org/10.1016/ j.clnu.2016.04.029
4
R.F. Tayyem et al. / Clinical Nutrition xxx (2016) 1e5
Table 4 Odds ratios and 95% confidence intervals for colorectal cancer by dietary patterns. Dietary pattern
Controls n (%)
Western Low 187 (66.5) High 94 (33.5) P-value Healthy Low 171 (60.9) High 110 (39.1) P-value High-sugar/high tea pattern Low 156 (55.5) High 125 (44.5) P-value
Cases n (%)
Odd ratioa
95%CI
109 (50.5) 111 (50.5)
1 1.88 0.001
1.12e3.16
127 (57.7) 93 (42.3)
1 0.93 0.480
0.56e1.53
126 (57.3) 94 (42.7)
1 0.8 0.695
0.51e1.27
a Adjusted for age, sex, total energy, MET minutes/week, tobacco use, education level, marital status, work, income, and family history of colorectal cancer, P < 0.05.
present study identified three dietary patterns among Jordanians, Western, Healthy and High Sugar/High tea patterns. Regarding the characteristics of our study group, 80.4% of the controls had BMI >25, that is higher than the result seen in the CRC group, 69.6%. The lower BMI for cases, as compared to controls, could be attributed to their cancer treatment. In the present study we found an association between Western dietary pattern and the risk of developing CRC. No associations were found for the healthy eating pattern or the High-Sugar/High Tea Pattern. The Western dietary pattern has been cited in previous studies [8,20] and this food choice pattern is generally characterized by high intake of meat and grains. Studies using factor analysis from both “Western” and other “developed” countries have linked the Western dietary pattern to the risk of developing CRC [16,20e22]. For example, Chen et al. (2015) reported a positive association between “meat dietary pattern” and CRC risk (OR ¼ 1.84; 95% CI ¼ 1.19e2.86) in a Canadian population study involving 506 CRC cases and 673 controls [20]. And in another cohort study of African American women [21], the Western dietary pattern was found to be associated with a 42% higher risk of colorectal adenoma [21]. Similarly, a case control study from Iran, conducted by Safri et al. (2013), also reported that a Western dietary pattern was associated with an increased risk of CRC development (OR ¼ 2.616; 95% CI ¼ 1.361e5.030) [16]. Additionally, De Stefani et al. (2011) also reported an increased risk of CRC development for colon cancer in Uruguayan men whose diet is similar to the western dietary pattern, OR of 2.62 (95% CI ¼ 1.36e5.08) [22]. In our study, both red meat and poultry were consumed in significant amounts in the described Jordanian Western dietary pattern. This is in agreement with Arafa et al. (2011) who reported that a diet high in red meat and saturated fat was associated with CRC among their studied Jordanian subjects [23]. The presence of poultry in this dietary pattern, has previously been reported by Aziz et al. (2015), who described this pattern as the “Iranian pattern” and this “Iranian Pattern” was also reported to be associated with the risk of developing CRC (OR ¼ 1.46; 95%CI ¼ 1.05e2.19) [24]. Poultry has been shown to be a good component in dietary pattern in Western countries. However, our study results confirm what was reported by Aziz et al. (2015). Perhaps, chicken should be included in the red meat category [24]. More importantly, we noted that the inclusion of chicken to the dietary Western pattern choice did not appear to change or influence the statistical outcome that showed an increased risk associated with the development of CRC. The potential mechanisms related to the consumption of food and the development of CRC is not clear, however, it is possible that more than one metabolic pathway, dietary compounds or biochemical reactions are involved. Current Knowledge suggests that carcinogenesis may be induced by meat group and other breakdown
products of metabolism such as heme iron, heterocyclic amines (HCAs), polycyclic aromatic hydrocarbons (PAHs), malondialdehyde, nitrites and nitrates [25]. On the other hand, total grains, sweets and deserts are also components of the Western dietary pattern; and the increased risk of CRC with consumption of grains and sweets and deserts may be due to the high glycemic index of these foods. Foods that induce hyperinsulinemia have been implicated in the etiology of CRC [26]. In this study we did not find a significant protective association between the healthy dietary pattern and CRC as was reported by Fung et al., 2003 [27]. Nevertheless, healthy dietary patterns were mostly loaded with fruits and vegetables. Fruits and vegetables are good source of components of fiber and folic acid, reported to have anti-carcinogenic effects [18]. And we have previously investigated and reported the lack of association of total fruit and vegetable intake with risk of CRC [28]. Van Duijnhoven suggested that the association of fruit and vegetables with CRC risk may be a reflection of increased intake of other food groups [29]. The healthy dietary pattern identified here also contained potatoes that earlier were identified in a report associated with increased risk of CRC in another study [6]. Our results have identified previously known dietary patterns with additional factors that may be relevant to our population. A unique pattern identified in this analysis is the High-Sugar/ High Tea Pattern. This pattern is mostly loaded with sugar and tea. Tea is a popular beverage in Jordan and it is served with sugar. We found no association of the High-Sugar/High Tea Pattern with the risk of CRC. Tea contains polyphenols that may protect against CRC [30]. 4.1. Study limitations First, selection and recall biases could influence the study results. Second, measuring dietary intakes using FFQ is susceptible to errors; however, we used a validated FFQ. Third, although we controlled for a wide range of potential confounders, in combination the three dietary patterns explained only 30.17% of the total variance in food intake. Finally, the sample size is small, and this may have an influence on the precision of the OR estimates. However, we enrolled available CRC patients from five large hospitals between January 2010 and December 2012. Jordan is a small country with the estimated population in 2009 of 5.98 million; hence, it was difficult to recruit more participants given that the overall age-standardized incidence of CRC was found to be 17.3 per million of population, in 2009. We accept that the availability of more participants would add weight to a more precise OR, nonetheless, we believe that our report has added to the discussion related to dietary patterns and colorectal cancer. In conclusion, our study identified three dietary patterns among Jordanians, Western, Healthy and High Sugar/High Tea Patterns. Consistent with previous studies, we found that a Western dietary pattern is associated with increased risk for CRC development. We did not see a protective effect in the healthy dietary pattern group, but this may have been due to the inclusion of other non-protective food items in this dietary pattern. The diet-disease risk relationship identified here could be used to promote healthy eating habits among our population. Statement of authorship Conception and design: Reema F. Tayyem. Development of methodology: Reema F. Tayyem, Hiba A. Bawadi. Acquisition of data: Ihab N. Shehadah, Tareq Al-Jaberi, Majed Al-Nusairr, Lana M. Agraib. Analysis and interpretation of data: Reema F. Tayyem, Suhad S. Abu-Mweis, Dennis D. Heath. Writing, review and/or
Please cite this article in press as: Tayyem RF, et al., Dietary patterns and colorectal cancer, Clinical Nutrition (2016), http://dx.doi.org/10.1016/ j.clnu.2016.04.029
R.F. Tayyem et al. / Clinical Nutrition xxx (2016) 1e5
revision of the manuscript: Reema F. Tayyem, Hiba A. Bawadi, Ihab N. Shehadah, Suhad S. Abu-Mweis, Dennis D. Heath, Tareq Al-Jaberi, Majed AlNusairr, Kamal E. Bani-Hani. Administrative, technical, or material support: Reema F. Tayyem, Kamal E. BaniHani. Study supervision: Reema F. Tayyem, Lana M. Agraib, Kamal E. Bani-Hani. Conflict of interest The authors declare that they have no conflict of interests. Acknowledgment The authors would like to thank the Higher Council of Science and Technology for sponsoring the research projects. Abbreviation CRC FFQ DHQ I PAR HEPA METs BMI
colorectal cancer food frequency questionnaire diet history questionnaire I 7-day physical activity recall health enhancing physical activity metabolic equivalents body mass index
References [1] World Cancer Research Fund/American Institute for Cancer Research. Food, nutrition, physical activity and the prevention of cancer: a global perspective. Washington, DC: American Institute for Cancer Research; 2007. [2] Heavy PM, McKenna D, Rowland IR. Colorectal cancer and the relationship between genes and the environment. Nutr Cancer 2004;48:124e41. [3] Ulrich CM, Bigler J, Whitton JA, Bostick R, Fosdick L, Potter JD. Epoxide hydrolase Tyr113His polymorphism is associated with elevated risk of colorectal polyps in the presence of smoking and high meat intake. Cancer Epidemiol Biomarkers Prev 2001;10:875e82. [4] Irigaray P, Newby JA, Clapp R, Hardell L, Howard V, Montagnier L, et al. Lifestyle-related factors and environmental agents causing cancer: an overview, Biomed Pharmacother 2007;61:640e58. [5] Sun Z, Zhu Y, Wang PP, Roebothan B, Zhao J, Dicks E, et al. Reported intake of selected micronutrients and risk of colorectal cancer: results from a large population-based caseecontrol study in Newfoundland, Labrador and Ontario, Canada. Anticancer Res 2012;32:687e96. [6] Flood A, Rastogi T, Wirfalt E, Mitrou PN, Reedy J, Subar AF, et al. Dietary patterns as identified by factor analysis and colorectal cancer among middleaged Americans. Am J Clin Nutr 2008;88:176e84. [7] Randi G, Edefonti V, Ferraroni M, La Vecchia C, Decarli A. Dietary patterns and the risk of colorectal cancer and adenomas. Nutr Rev 2010;68:389e408. [8] Magalhaes B, Bastos J, Lunet N. Dietary patterns and colorectal cancer: a caseecontrol study from Portugal. Eur J Cancer Prev 2011;20:389e95. [9] Magalhaes B, Peleteiro B, Lunet N. Dietary patterns and colorectal cancer: systematic review and meta-analysis. Eur J Cancer Prev 2012;21:15e23.
5
[10] Tayyem RF, Bawadi H, Shehadeh I, Abu-Mweis SS, Agraib L, Bani-Hani K, et al. Macro- and micronutrients consumption and the risk for colorectal cancer among Jordanians. Nutrients 2015;7(3):1769e86. [11] Tayyem RF, Bawadi H, Shehadeh I, Agraib L, Bani-Hani K, Al-Awwad N, et al. Consumption of whole grains, refined cereals, and legumes and its association with colorectal cancer among Jordanians. Integr Cancer Ther 2016. http:// dx.doi.org/10.1177/1534735415620010. [12] Tayyem RF, Abu-Mweis SS, Bawadi HA, Agraib L, Bani-Hani K. Validation of a food frequency questionnaire to assess macronutrient and micronutrient intake among Jordanians. J Acad Nutr Diet 2014;114:1046e52. [13] Sallis JF, Haskell WL, Wood PD. Physical activity assessment methodology in the five-city project. Am J Epidemiol 1985;121:91e106. [14] Lee RD, Nieman DC. Nutritional assessment. 6th ed. New York: McGraw Hill; 2012. [15] Hajizadeh B, Rashidkhani B, Rad AH, Moasheri SM, Saboori H. Dietary patterns and risk of esophageal squamous cell carcinoma: a caseecontrol study. Public Health Nutr 2010;13(7):1107e12. [16] Safari A, Shariff ZM, Kandiah M, Rashidkhani B, Fereidooni F. Dietary patterns and risk of colorectal cancer in Tehran Province: a caseecontrol study. BMC Public Health 2013;13:222. [17] Hu FB, Rimm E, Smith-Warner SA, Feskanich D, Stampfer MJ, Ascherio A, et al. Reproducibility and validity of dietary patterns assessed with a foodfrequency questionnaire. Am J Clin Nutr 1999;69:243e9. [18] Harshman MR, Aldoori W. Diet and colorectal cancer: review of the evidence. Can Fam Physician 2007;53:1913e20. [19] Chao A, Thun MJ, Jacobs EJ, Henley SJ, Rodriguez C, Calle EE. Cigarette smoking and colorectal cancer mortality in the cancer prevention study II. J Natl Cancer Inst 2000;92:1888e96. [20] Chen Z, Wang PP, Woodrow J, Zhu Y, Roebothan B, Mclaughlin JR, et al. Dietary patterns and colorectal cancer: results from a Canadian population-based study. Nutr J 2015;15(14):8. [21] Makambi KH, Agurs-Collins T, Bright-Gbebry M, Rosenberg L, Palmer JR, Adams-Campbell LL. Dietary patterns and the risk of colorectal adenomas: the Black Women's Health Study. Cancer Epidemiol Biomarkers Prev 2011;20(5): 818e25. [22] Stefani ED, Deneo-Pellegrini H, Ronco AL, Correa P, Boffetta P, Aune D, et al. Dietary patterns and risk of colorectal cancer: a factor analysis in Uruguay. Asian Pac J Cancer Prev 2011;12(3):753e9. [23] Arafa MA, Waly MI, Jriesat S, Al Khafajei A, Sallam S. Dietary and lifestyle characteristics of colorectal cancer in Jordan: a case-control study. Asian Pac J Cancer Prev 2011;12(8):1931e6. [24] Azizi H, Asadollahi K, Davtalab Esmaeili E, Mirzapoor M. Iranian dietary patterns and risk of colorectal cancer. Health Promot Perspect 2015;5(1):72e80. [25] Durko L, Malecka-Panas E. Lifestyle modifications and colorectal cancer. Curr Colorectal Cancer Rep 2014;10:45e54. [26] Giovannucci E. Insulin, insulin-like growth factors and colon cancer: a review of the evidence. J Nutr 2001;131:3109Se20S. [27] Fung T, Hu FB, Fuchs C, Giovannucci E, Hunter DJ, Stampfer MJ, et al. Major dietary patterns and the risk of colorectal cancer in women. Arch Intern Med 2003;163(3):309e14. [28] Tayyem RF, Shehadeh I, Abu-Mweis SS, Bawadi H, Bani-Hani K, Al-Jaberi T, et al. Fruit and vegetable intake among Jordanians: results from a case-control study of colorectal cancer. Cancer Control J Moffitt Cancer Cent 2014;21(4): 350e6. [29] van Duijnhoven FJ, Bueno-De-Mesquita HB, Ferrari P, Jenab M, Boshuizen HC, Ros MM, et al. Fruit, vegetables, and colorectal cancer risk: the European prospective investigation into cancer and nutrition. Am J Clin Nutr 2009;89: 1441e52. [30] Siddiqui IA, Afaq F, Adhami VM, Ahmad N, Mukhtar H. Antioxidants of the beverage tea in promotion of human health. Antioxid Redox Signal 2004;6(3): 571e82.
Please cite this article in press as: Tayyem RF, et al., Dietary patterns and colorectal cancer, Clinical Nutrition (2016), http://dx.doi.org/10.1016/ j.clnu.2016.04.029
Contents lists available at ScienceDirect
Clinical Nutrition journal homepage: http://www.elsevier.com/locate/clnu
Original article
Dietary patterns and colorectal cancer Reema F. Tayyem a, *, Hiba A. Bawadi b, Ihab Shehadah c, Lana M. Agraib d, Suhad S. AbuMweis d, Tareq Al-Jaberi e, Majed Al-Nusairr f, Kamal E. Bani-Hani g, Dennis D. Heath h a
Department of Nutrition and Food Technology, Faculty of Agriculture, The University of Jordan, Amman 11942 Jordan Human Nutrition Department, College of Health Sciences, Qatar University, P.O. Box: 2713, Doha, Qatar c Chief Gastroenterology Division, King Hussein Cancer Center, Jordan d Department of Clinical Nutrition & Dietetic, The Hashemite University, P.O. Box 150459, Zarqa 13115, Jordan e Jordan University of Science and Technology, Jordan f Chief Gastroenterology Division, Prince Hamza Hospital, Jordan g Faculty of Medicine, Hashemite University, Jordan h Cancer Prevention and Control Program, Moores Cancer Center, University of California, San Diego, La Jolla, CA 92093, USA b
a r t i c l e i n f o
s u m m a r y
Article history: Received 8 February 2016 Accepted 28 April 2016
Background & aims: Dietary pattern and lifestyle have been reported to be important risk factors in the development of colorectal cancer (CRC). However, the mechanism of action of dietary factors in CRC disease is unclear. The aim of this study is the examination of several dietary choices and their potential association with the risk of developing CRC. Methods: Dietary data was collected from 220 subjects who were previously diagnosed with CRC, and 281 control subjects (matched by age, gender, occupation and marital status). The data was collected between January 2010 and December 2012, using interview-based questionnaires. Multivariate logistic regression was used to estimate the relationship between dietary choices and risk of developing colorectal cancer. Results: Factor analysis revealed three major dietary patterns. The first pattern we identified as the “Healthy Pattern”, the second was identified as “High Sugar/High Tea Pattern” and the third as “Western Pattern”. In the Healthy Pattern group we found a 10.54% variation in food intake, while the intake variation was 11.64% in the Western Pattern. After adjusting for confounding factors, the Western Pattern food choice was found to be significantly associated with an increased risk of developing CRC (OR ¼ 1.88; 95% CI ¼ 1.12e3.16). The results for the Healthy and High-Sugar/High Tea Patterns showed a decrease, but the statistic was not significant for the risk of CRC development. Conclusion: The Western Pattern of dietary choice was directly associated with CRC. The association between the dietary food choice in the Healthy and High-Sugar/High Tea Patterns and colorectal cancer needs further study in our Jordanian population. © 2016 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.
Keywords: Colorectal cancer Dietary pattern Healthy dietary pattern Western dietary pattern
1. Introduction In colorectal cancer (CRC) disease, several well-known dietary and non-dietary risk factors have been implicated. Some of those factors are high consumption of red meat and processed meat; low
* Corresponding author. E-mail addresses: [email protected] (R.F. Tayyem), [email protected] (H.A. Bawadi), [email protected] (I. Shehadah), [email protected] (L.M. Agraib), [email protected] (S.S. AbuMweis), tmrjaberi@hotmail. com (T. Al-Jaberi), [email protected] (K.E. Bani-Hani), [email protected] (D.D. Heath).
fiber intake; alcohol drinking; obesity; and a sedentary life style [1]. Additionally, genetic susceptibility [2], tobacco smoking [3], and exposure to environmental carcinogens, were found to promote proliferation and malignant transformation of CRC cells [4]. Previous studies have focused on the effects of a single food item or a nutrient on lowering risk of CRC incidence [5]. However, the association of a single food item or food group with the risk of developing CRC may not be valid because of the presumption that each single food item or nutrient has an isolated effect [6]. A dietary pattern in food choice is defined as a combination of the dietary components (food items, food groups, nutrients, or both) used to summarize elements of the total diet or the major features of the
http://dx.doi.org/10.1016/j.clnu.2016.04.029 0261-5614/© 2016 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.
Please cite this article in press as: Tayyem RF, et al., Dietary patterns and colorectal cancer, Clinical Nutrition (2016), http://dx.doi.org/10.1016/ j.clnu.2016.04.029
2
R.F. Tayyem et al. / Clinical Nutrition xxx (2016) 1e5
food choices for the population under study [7]. The descriptive summary of the dietary pattern has been used in nutritional epidemiology to explain and assess the overall dietary experience, by suggesting that the synergistic effects of the variety of dietary and non-dietary factors can be used to explain the relationship between diet and health [8]. In general there are two dietary patterns; “Healthy” and “Western”. The healthy dietary pattern is largely characterized by a greater intake of fruits, vegetables, grains, and a lower intake of sweets, red meat, and processed meat, this dietary pattern is considered to be associated with lowering the risk of developing CRC. Alternatively, the Western dietary pattern, reported to contain more meat, highly processed food, potatoes, refined carbohydrates, and much lower in vegetables and dietary fiber, has been reported to increase the risk of developing CRC [7,9]. The numbers of studies that have examined the possible effect of dietary patterns and CRC disease in the developing countries are few. However, we previously reported on the association between macro- and micronutrients consumption and the risk for colorectal cancer among Jordanians [10]. In this study we aimed to examine the major dietary patterns and their possible association in the development of CRC in a Jordanian population, using factor analysis or Principal Component Analysis. 2. Materials and methods 2.1. Study population and methods A detailed description of the method has been reported elsewhere [11]. Briefly, a case-control study of 501 participants was conducted from January 2010 to December 2012; with 220 diagnosed with CRC disease, and 281 controls (a total of 248 males and 253 females). Patients diagnosed with CRC disease (cases) were recruited from five Jordanian hospitals specializing in oncology diagnosis and treatment. The Control group was recruited from outpatients, hospital personnel, and visitors. Controls were matched with cases for age, gender, occupation and marital status. The study protocol was approved by Institutional Review Board Committees from all hospital groups. Control subjects were excluded if any first- or second degree relatives were diagnosed with CRC. The following inclusion criteria for controls were used: Jordanian nationals aged 18 years or older, ability to communicate clearly and verbally, free of any type of diagnosed cancer, diabetes mellitus, liver disease and rheumatoid arthritis. For inclusion in the diagnosed CRC cancer group, subjects must have received their diagnosis less than 1 year prior to the time of the first interview. The exclusion criteria for this group included those who were considered “critically ill”, such as an in-patient at any facility and those who were unable to communicate verbally and clearly. Written informed consent was obtained from all subjects before their interview. 2.2. Data collection Socio-demographic information (age, marital status, household income, education, occupation and tobacco usage); medical history as well as dietary data were collected by trained research assistants using interview-based questionnaires. The comprehensive medical data included the participant's full medical history to confirm the status of CRC diagnosed subjects and disease-free subjects. A validated Arabic quantitative Food Frequency Questionnaire (FFQ), adapted from the Diet History Questionnaire (DHQ I) of the National Cancer Institute of the United States of America [12] was used for dietary assessment. The FFQ questions tracked the information on the dietary history of study participants prior to CRC
diagnosis, and to confirm the dietary habits of control participants. A period of one year prior to the diagnosis date was selected, to reflect seasonal variation in some food types. A constant dietary pattern for the period was noted in our participants, with most of the participants suggesting this pattern existed for at least five years. A qualified dietitian asked participants, during face-to-face interviews, how frequently, on average, during the past year they had consumed one standard serving of specific food items in nine categories (<1/month, 2e3/month, 1e2/week, 3e4/week, 5e6/ week, 1/day, 2e3/day, 4e5/day, or 6/day). An answer in the affirmative resulted in additional questions related to frequency and amount of food consumed. If the participants' dietary pattern did not include a food type, then related questions were skipped. Food lists in the modified FFQ questions were classified based on types of foods: 21 items of fruits and juices; 21 items of vegetables; eight items of cereals; nine items of milk and dairy products; four items of beans; 16 items of meat such as red meat (lamb and beef), chicken, fish, cold meat, and others; four items of soups and sauces; five items of drinks; nine items of snacks and sweets; and 14 items of herbs and spices [12]. For better portion size estimation food models and standard measuring tools were used. Dietary intakes were analyzed using dietary analysis software (ESHA Food Processor SQL version 10.1.1; ESHA, Salem, OR, USA) with additional data on foods consumed in Jordan. The 7-day Physical Activity Recall (PAR), developed by Sallis et al. (1985) was used to measure physical activity level. 7-Day PAR is a structured interview that depends on participant's recall of time spent engaging in physical activity over a seven day period [13]. Participants were asked to respond to a PAR question based on the way they used to behave prior to being diagnosed with CRC. The number of hours spent in different activity levels were obtained and converted into metabolic equivalents (METs). The total physical activity MET minutes per week was obtained by summing the METs and then performing categorical analysis (inactive, minimally active, or health enhancing physical activity active) [13]. Body weight (measured to the nearest 0.1 kg); height (measured to the nearest 0.5 cm); and body mass index (BMI) were taken and calculated as prescribed by Lee and Nieman [14]. 2.3. Statistical analyses Statistical analysis was performed with SPSS IBM-20 software. The significance level was set at p 0.05. For descriptive statistics, mean ± standard deviation (SD) and percentages were used. T-tests evaluated the differences between cases and controls in continuous variables, and Chi-square was used to detect differences among categorical variables. Dietary patterns were derived using Principal Component Analysis (PCA), form factor analysis. The food items in the FFQ were separated into 22 food groups, based on their similarity of nutrient content and culinary usage or their reported relationship with cancer [15] (Table 1). Kaiser-Meyer-Olkin (KMOtest) and Bartlett's test of sphericity were used to assess suitability for using factor analysis for this exercise. As suggested by Safari et al. (2013) [16] sampling adequacy and inter-correlation of factors were supported by KMO value > 0.632 and Bartlett's test of sphericity < 0.001, respectively. Communality index was assessed to indicate the variance in each food group being tested by the analysis [16]. Factors were retained based on an eigenvalue of >1 for screen plot. Then Varimax rotation was applied to review the correlations between variables and factors [16]. Food groups with absolute factor loadings >0.20 were considered to have contributed significantly to the pattern. Cases and controls received an individual factor score for each identified pattern [17]. Potential confounders (with age, sex, BMI, physical activity level (MET-min/ week), total energy intake, occupation, education level, smoking,
Please cite this article in press as: Tayyem RF, et al., Dietary patterns and colorectal cancer, Clinical Nutrition (2016), http://dx.doi.org/10.1016/ j.clnu.2016.04.029
R.F. Tayyem et al. / Clinical Nutrition xxx (2016) 1e5 Table 1 Food groups used in the dietary pattern analysis. Food group
3
Table 2 Selected characteristics of the participants' study. (Tayyem et al., 2016).
Food items
Processed meat Red Meat Fish Poultry Egg Dairies Tea Coffee Fruits
Sausages, hamburger, mortadella, turkey Beef meat, ground meat, lamp meat Tuna, any type of fish Chicken Fried eggs, boiled eggs Milk, yogurt, cheese, ice cream, labaneh. Black tea, green tea Coffee Watermelon, melon, apple, apricot, fig, nectarine, peach, pear, Citrus fruit, date, kiwi, grape, strawberry, banana, grape fruit, alovera, other fruits Tomato Tomato Carrot Carrot vegetables Spinach, lettuce, mixed vegetable, stew vegetables, eggplant, green squash, local vegetables, pepper, cucumber, garlic, kinds of cabbage, root vegetables, other vegetables Legumes Bean, chickpea, split pea, lentil, other cereals Fried potato Fried potato Boiled potato Boiled potato Total grains Bread, rice, macaroni Snacks Biscuits, puff, chips Nuts Peanut, almond, walnut, pistachio, hazelnut, roasted seeds Sweets and desert Cakes, cookies, chocolate, Arabic sweet, candy Sugar Sugar, sugar cube Mayonnaise Mayonnaise Soft drink Carbonated drinks
marital status and family history (beyond the second degree) for the CRC participants) were chosen based on reported risk factors for CRC [18,19], including the Cancer Prevention Study II [19]. 3. Results Cases and controls were matched for several parameters, including age, sex, occupation and marital status. Characteristics specific to our subjects were published previously [11] and shown in Table 2. The mean age for controls and cases was 51.3 ± 11.1 and 52.9 ± 11.7 respectively, while mean BMI was 29.1 ± 5.6 and 27.7 ± 6.2 for controls and cases, respectively. The Factor loading matrix analysis showed three major dietary patterns. These three distinct dietary patterns, labeled “Western Pattern”, “Healthy Pattern” and “High-Sugar/High Tea Pattern”, were extracted using the above-mentioned factor analysis procedure. The three dietary patterns explained 32.26% of the total variance in food intake. The first pattern was heavily loaded with processed and red meat, total grains, coffee, fried potato, nuts, poultry, sweets and desserts. This group of food was identified as the “Western” dietary pattern. The Western dietary pattern was responsible for 11.64% of the total variance. The second pattern, labeled “Healthy” dietary pattern, had a high amount of fruits, vegetables, dairies, tomato, carrot, and nuts. While, the third pattern identified as “High-Sugar/High Tea Pattern”, had food group that was significant due to the higher amounts of sugar and tea, and obtained results had a variance of 7.99%. The cumulative variance was about 30% (Table 3). After adjusting for confounding factors, an increased risk of developing CRC was positively associated with the Western dietary pattern (OR ¼ 1.88; 95% CI ¼ 1.12e3.16), while an insignificant decreased risk of developing CRC was observed in the Healthy Pattern food choice group (OR ¼ 0.93; 95% CI ¼ 0.56e1.53) as shown in Table 4. 4. Discussions This case-control study investigated several dietary patterns and their potential association with the risk of developing CRC. The
Age (mean ± SD) Sex n (%) Male Female BMI (mean ± SD) BMI Category Under weight (<18.5) Normal (18.5e24.9) Over weight (25e29.9) Obese 30 Marital status n (%) Married Single Divorced Widowed Occupation n (%) Yes No Smoking n (%) Smoker Non-smoker Family history of CRC n (%) Yes No Other health problem n (%) Yes No Education n (%) Illiterate primary and secondary Diploma and BSc MSc and PhD MET n (%) Inactivea Minimally Activeb HEPA activec
Control (n ¼ 281)
Case (n ¼ 220)
51.25 ± 11.12
52.89 ± 11.67
132 (47) 149 (53) 29.06 ± 5.6
116 (52.7) 104 (47.3) 27.71 ± 6.15
2 (0.7) 52 (18.8) 122 (44.2) 100 (36.2)
5 (2.5) 57 (27.9) 83 (40.7) 59 (28.9)
248 (88.3) 17 (6) 1 (0.4) 15 (5.3)
199 (90.5) 5 (2.3) 3 (1.4) 13 (5.9)
100 (35.6) 181 (64.4)
69 (31.4) 151 (68.6)
53 (18.9) 227 (81.1)
37 (16.8) 173 (78.6)
101 (36.5) 176 (63.5)
84 (38.5) 134 (61.5)
119 (42.7) 160 (57.3)
83 (37.7) 136 (61.8)
11 (3.9) 137 (49.1) 113 (40.5) 18 (6.5)
17 (7.7) 107 (56.4) 86 (39.1) 10 (4.5)
131 (52.2) 53 (21.1) 67 (26.7)
121 (56.3) 37 (17.2) 57 (26.5)
BMI: Body Mass Index. a Inactive: not fitting in “Minimally Active” or “HEPA active”. b Minimally Active: at least 600 MET per week. c Health Enhancing Physical Activity: HEPA active: more than 3000 MET per week.
Table 3 Factor loading matrix of food groups for Healthy and Western dietary patterns. Food group Egg Total grains Red meat Fish Coffee Soft drink Processed meat Sweets and desert Mayonnaise Nuts Snacks Chicken Legumes Fruits Carrots Vegetables Tomato Dairies Fried potatoes Boiled potatoes Tea Sugar Total variance
Western Pattern 0.512 0.491 0.490 0.451 0.449 0.431 0.428 0.400 0.388 0.388 0.369 0.350 0.292
0.212 0.335
11.64%
Healthy Pattern
High-sugar/ high tea pattern
0.274
.237 0.212 0.264
0.655 0.652 0.650 0.556 0.441 0.370 0.306
10.54%
0.240 0.231 0.865 0.836 7.99%
Absolute factor loading values <0.20 for the three patterns were excluded for simplicity. Bold values represent foods with highest contribution to each patterns.
Please cite this article in press as: Tayyem RF, et al., Dietary patterns and colorectal cancer, Clinical Nutrition (2016), http://dx.doi.org/10.1016/ j.clnu.2016.04.029
4
R.F. Tayyem et al. / Clinical Nutrition xxx (2016) 1e5
Table 4 Odds ratios and 95% confidence intervals for colorectal cancer by dietary patterns. Dietary pattern
Controls n (%)
Western Low 187 (66.5) High 94 (33.5) P-value Healthy Low 171 (60.9) High 110 (39.1) P-value High-sugar/high tea pattern Low 156 (55.5) High 125 (44.5) P-value
Cases n (%)
Odd ratioa
95%CI
109 (50.5) 111 (50.5)
1 1.88 0.001
1.12e3.16
127 (57.7) 93 (42.3)
1 0.93 0.480
0.56e1.53
126 (57.3) 94 (42.7)
1 0.8 0.695
0.51e1.27
a Adjusted for age, sex, total energy, MET minutes/week, tobacco use, education level, marital status, work, income, and family history of colorectal cancer, P < 0.05.
present study identified three dietary patterns among Jordanians, Western, Healthy and High Sugar/High tea patterns. Regarding the characteristics of our study group, 80.4% of the controls had BMI >25, that is higher than the result seen in the CRC group, 69.6%. The lower BMI for cases, as compared to controls, could be attributed to their cancer treatment. In the present study we found an association between Western dietary pattern and the risk of developing CRC. No associations were found for the healthy eating pattern or the High-Sugar/High Tea Pattern. The Western dietary pattern has been cited in previous studies [8,20] and this food choice pattern is generally characterized by high intake of meat and grains. Studies using factor analysis from both “Western” and other “developed” countries have linked the Western dietary pattern to the risk of developing CRC [16,20e22]. For example, Chen et al. (2015) reported a positive association between “meat dietary pattern” and CRC risk (OR ¼ 1.84; 95% CI ¼ 1.19e2.86) in a Canadian population study involving 506 CRC cases and 673 controls [20]. And in another cohort study of African American women [21], the Western dietary pattern was found to be associated with a 42% higher risk of colorectal adenoma [21]. Similarly, a case control study from Iran, conducted by Safri et al. (2013), also reported that a Western dietary pattern was associated with an increased risk of CRC development (OR ¼ 2.616; 95% CI ¼ 1.361e5.030) [16]. Additionally, De Stefani et al. (2011) also reported an increased risk of CRC development for colon cancer in Uruguayan men whose diet is similar to the western dietary pattern, OR of 2.62 (95% CI ¼ 1.36e5.08) [22]. In our study, both red meat and poultry were consumed in significant amounts in the described Jordanian Western dietary pattern. This is in agreement with Arafa et al. (2011) who reported that a diet high in red meat and saturated fat was associated with CRC among their studied Jordanian subjects [23]. The presence of poultry in this dietary pattern, has previously been reported by Aziz et al. (2015), who described this pattern as the “Iranian pattern” and this “Iranian Pattern” was also reported to be associated with the risk of developing CRC (OR ¼ 1.46; 95%CI ¼ 1.05e2.19) [24]. Poultry has been shown to be a good component in dietary pattern in Western countries. However, our study results confirm what was reported by Aziz et al. (2015). Perhaps, chicken should be included in the red meat category [24]. More importantly, we noted that the inclusion of chicken to the dietary Western pattern choice did not appear to change or influence the statistical outcome that showed an increased risk associated with the development of CRC. The potential mechanisms related to the consumption of food and the development of CRC is not clear, however, it is possible that more than one metabolic pathway, dietary compounds or biochemical reactions are involved. Current Knowledge suggests that carcinogenesis may be induced by meat group and other breakdown
products of metabolism such as heme iron, heterocyclic amines (HCAs), polycyclic aromatic hydrocarbons (PAHs), malondialdehyde, nitrites and nitrates [25]. On the other hand, total grains, sweets and deserts are also components of the Western dietary pattern; and the increased risk of CRC with consumption of grains and sweets and deserts may be due to the high glycemic index of these foods. Foods that induce hyperinsulinemia have been implicated in the etiology of CRC [26]. In this study we did not find a significant protective association between the healthy dietary pattern and CRC as was reported by Fung et al., 2003 [27]. Nevertheless, healthy dietary patterns were mostly loaded with fruits and vegetables. Fruits and vegetables are good source of components of fiber and folic acid, reported to have anti-carcinogenic effects [18]. And we have previously investigated and reported the lack of association of total fruit and vegetable intake with risk of CRC [28]. Van Duijnhoven suggested that the association of fruit and vegetables with CRC risk may be a reflection of increased intake of other food groups [29]. The healthy dietary pattern identified here also contained potatoes that earlier were identified in a report associated with increased risk of CRC in another study [6]. Our results have identified previously known dietary patterns with additional factors that may be relevant to our population. A unique pattern identified in this analysis is the High-Sugar/ High Tea Pattern. This pattern is mostly loaded with sugar and tea. Tea is a popular beverage in Jordan and it is served with sugar. We found no association of the High-Sugar/High Tea Pattern with the risk of CRC. Tea contains polyphenols that may protect against CRC [30]. 4.1. Study limitations First, selection and recall biases could influence the study results. Second, measuring dietary intakes using FFQ is susceptible to errors; however, we used a validated FFQ. Third, although we controlled for a wide range of potential confounders, in combination the three dietary patterns explained only 30.17% of the total variance in food intake. Finally, the sample size is small, and this may have an influence on the precision of the OR estimates. However, we enrolled available CRC patients from five large hospitals between January 2010 and December 2012. Jordan is a small country with the estimated population in 2009 of 5.98 million; hence, it was difficult to recruit more participants given that the overall age-standardized incidence of CRC was found to be 17.3 per million of population, in 2009. We accept that the availability of more participants would add weight to a more precise OR, nonetheless, we believe that our report has added to the discussion related to dietary patterns and colorectal cancer. In conclusion, our study identified three dietary patterns among Jordanians, Western, Healthy and High Sugar/High Tea Patterns. Consistent with previous studies, we found that a Western dietary pattern is associated with increased risk for CRC development. We did not see a protective effect in the healthy dietary pattern group, but this may have been due to the inclusion of other non-protective food items in this dietary pattern. The diet-disease risk relationship identified here could be used to promote healthy eating habits among our population. Statement of authorship Conception and design: Reema F. Tayyem. Development of methodology: Reema F. Tayyem, Hiba A. Bawadi. Acquisition of data: Ihab N. Shehadah, Tareq Al-Jaberi, Majed Al-Nusairr, Lana M. Agraib. Analysis and interpretation of data: Reema F. Tayyem, Suhad S. Abu-Mweis, Dennis D. Heath. Writing, review and/or
Please cite this article in press as: Tayyem RF, et al., Dietary patterns and colorectal cancer, Clinical Nutrition (2016), http://dx.doi.org/10.1016/ j.clnu.2016.04.029
R.F. Tayyem et al. / Clinical Nutrition xxx (2016) 1e5
revision of the manuscript: Reema F. Tayyem, Hiba A. Bawadi, Ihab N. Shehadah, Suhad S. Abu-Mweis, Dennis D. Heath, Tareq Al-Jaberi, Majed AlNusairr, Kamal E. Bani-Hani. Administrative, technical, or material support: Reema F. Tayyem, Kamal E. BaniHani. Study supervision: Reema F. Tayyem, Lana M. Agraib, Kamal E. Bani-Hani. Conflict of interest The authors declare that they have no conflict of interests. Acknowledgment The authors would like to thank the Higher Council of Science and Technology for sponsoring the research projects. Abbreviation CRC FFQ DHQ I PAR HEPA METs BMI
colorectal cancer food frequency questionnaire diet history questionnaire I 7-day physical activity recall health enhancing physical activity metabolic equivalents body mass index
References [1] World Cancer Research Fund/American Institute for Cancer Research. Food, nutrition, physical activity and the prevention of cancer: a global perspective. Washington, DC: American Institute for Cancer Research; 2007. [2] Heavy PM, McKenna D, Rowland IR. Colorectal cancer and the relationship between genes and the environment. Nutr Cancer 2004;48:124e41. [3] Ulrich CM, Bigler J, Whitton JA, Bostick R, Fosdick L, Potter JD. Epoxide hydrolase Tyr113His polymorphism is associated with elevated risk of colorectal polyps in the presence of smoking and high meat intake. Cancer Epidemiol Biomarkers Prev 2001;10:875e82. [4] Irigaray P, Newby JA, Clapp R, Hardell L, Howard V, Montagnier L, et al. Lifestyle-related factors and environmental agents causing cancer: an overview, Biomed Pharmacother 2007;61:640e58. [5] Sun Z, Zhu Y, Wang PP, Roebothan B, Zhao J, Dicks E, et al. Reported intake of selected micronutrients and risk of colorectal cancer: results from a large population-based caseecontrol study in Newfoundland, Labrador and Ontario, Canada. Anticancer Res 2012;32:687e96. [6] Flood A, Rastogi T, Wirfalt E, Mitrou PN, Reedy J, Subar AF, et al. Dietary patterns as identified by factor analysis and colorectal cancer among middleaged Americans. Am J Clin Nutr 2008;88:176e84. [7] Randi G, Edefonti V, Ferraroni M, La Vecchia C, Decarli A. Dietary patterns and the risk of colorectal cancer and adenomas. Nutr Rev 2010;68:389e408. [8] Magalhaes B, Bastos J, Lunet N. Dietary patterns and colorectal cancer: a caseecontrol study from Portugal. Eur J Cancer Prev 2011;20:389e95. [9] Magalhaes B, Peleteiro B, Lunet N. Dietary patterns and colorectal cancer: systematic review and meta-analysis. Eur J Cancer Prev 2012;21:15e23.
5
[10] Tayyem RF, Bawadi H, Shehadeh I, Abu-Mweis SS, Agraib L, Bani-Hani K, et al. Macro- and micronutrients consumption and the risk for colorectal cancer among Jordanians. Nutrients 2015;7(3):1769e86. [11] Tayyem RF, Bawadi H, Shehadeh I, Agraib L, Bani-Hani K, Al-Awwad N, et al. Consumption of whole grains, refined cereals, and legumes and its association with colorectal cancer among Jordanians. Integr Cancer Ther 2016. http:// dx.doi.org/10.1177/1534735415620010. [12] Tayyem RF, Abu-Mweis SS, Bawadi HA, Agraib L, Bani-Hani K. Validation of a food frequency questionnaire to assess macronutrient and micronutrient intake among Jordanians. J Acad Nutr Diet 2014;114:1046e52. [13] Sallis JF, Haskell WL, Wood PD. Physical activity assessment methodology in the five-city project. Am J Epidemiol 1985;121:91e106. [14] Lee RD, Nieman DC. Nutritional assessment. 6th ed. New York: McGraw Hill; 2012. [15] Hajizadeh B, Rashidkhani B, Rad AH, Moasheri SM, Saboori H. Dietary patterns and risk of esophageal squamous cell carcinoma: a caseecontrol study. Public Health Nutr 2010;13(7):1107e12. [16] Safari A, Shariff ZM, Kandiah M, Rashidkhani B, Fereidooni F. Dietary patterns and risk of colorectal cancer in Tehran Province: a caseecontrol study. BMC Public Health 2013;13:222. [17] Hu FB, Rimm E, Smith-Warner SA, Feskanich D, Stampfer MJ, Ascherio A, et al. Reproducibility and validity of dietary patterns assessed with a foodfrequency questionnaire. Am J Clin Nutr 1999;69:243e9. [18] Harshman MR, Aldoori W. Diet and colorectal cancer: review of the evidence. Can Fam Physician 2007;53:1913e20. [19] Chao A, Thun MJ, Jacobs EJ, Henley SJ, Rodriguez C, Calle EE. Cigarette smoking and colorectal cancer mortality in the cancer prevention study II. J Natl Cancer Inst 2000;92:1888e96. [20] Chen Z, Wang PP, Woodrow J, Zhu Y, Roebothan B, Mclaughlin JR, et al. Dietary patterns and colorectal cancer: results from a Canadian population-based study. Nutr J 2015;15(14):8. [21] Makambi KH, Agurs-Collins T, Bright-Gbebry M, Rosenberg L, Palmer JR, Adams-Campbell LL. Dietary patterns and the risk of colorectal adenomas: the Black Women's Health Study. Cancer Epidemiol Biomarkers Prev 2011;20(5): 818e25. [22] Stefani ED, Deneo-Pellegrini H, Ronco AL, Correa P, Boffetta P, Aune D, et al. Dietary patterns and risk of colorectal cancer: a factor analysis in Uruguay. Asian Pac J Cancer Prev 2011;12(3):753e9. [23] Arafa MA, Waly MI, Jriesat S, Al Khafajei A, Sallam S. Dietary and lifestyle characteristics of colorectal cancer in Jordan: a case-control study. Asian Pac J Cancer Prev 2011;12(8):1931e6. [24] Azizi H, Asadollahi K, Davtalab Esmaeili E, Mirzapoor M. Iranian dietary patterns and risk of colorectal cancer. Health Promot Perspect 2015;5(1):72e80. [25] Durko L, Malecka-Panas E. Lifestyle modifications and colorectal cancer. Curr Colorectal Cancer Rep 2014;10:45e54. [26] Giovannucci E. Insulin, insulin-like growth factors and colon cancer: a review of the evidence. J Nutr 2001;131:3109Se20S. [27] Fung T, Hu FB, Fuchs C, Giovannucci E, Hunter DJ, Stampfer MJ, et al. Major dietary patterns and the risk of colorectal cancer in women. Arch Intern Med 2003;163(3):309e14. [28] Tayyem RF, Shehadeh I, Abu-Mweis SS, Bawadi H, Bani-Hani K, Al-Jaberi T, et al. Fruit and vegetable intake among Jordanians: results from a case-control study of colorectal cancer. Cancer Control J Moffitt Cancer Cent 2014;21(4): 350e6. [29] van Duijnhoven FJ, Bueno-De-Mesquita HB, Ferrari P, Jenab M, Boshuizen HC, Ros MM, et al. Fruit, vegetables, and colorectal cancer risk: the European prospective investigation into cancer and nutrition. Am J Clin Nutr 2009;89: 1441e52. [30] Siddiqui IA, Afaq F, Adhami VM, Ahmad N, Mukhtar H. Antioxidants of the beverage tea in promotion of human health. Antioxid Redox Signal 2004;6(3): 571e82.
Please cite this article in press as: Tayyem RF, et al., Dietary patterns and colorectal cancer, Clinical Nutrition (2016), http://dx.doi.org/10.1016/ j.clnu.2016.04.029